death to the peak limiter!
3 Likes
And yet going to (most) live music events without industrial hearing protection results in permanent hearing loss…maybe Internet is not the only problem?
4 Likes
If you cant beat them, join them start your own streaming service.
2 Likes
Well I hope Neil Young will remember, the internet don’t need him around anyhow…
8 Likes
In general he’s wrong, but with one exception – just have a listen to today’s satellite radio “music”. Oh my GOD it’s a hideous swishing swirl of overcompressed 32kbps bitmush. So so bad.
I am guessing digital music streaming is probably ~128kbps these days and with a modern codec that’s pretty solid. 192kbps is where it starts to be indistinguishable from original source, assuming it was encoded sanely in the first place of course.
https://support.spotify.com/us/article/high-quality-streaming/
Ah, looks like the high quality setting for free is 160kbps which is quite good. Normal is only 96kbps though. Premium subscription unlocks a 320kbps “very high” which is borderline archival quality.
Web streaming is 128kbps as I suspected.
4 Likes
I listen to a lot of music through an almost cartoonishly terrible signal path: phone via bluetooth to an FM transmitter which sends to my car radio. Contrary to Mr. Young’s thesis, it makes me appreciate the nuances I can pick out when listening to the same music on better equipment all the more.
5 Likes
This sound engineer disagrees. Mastering ears are absolutely next level as are many recording artists who are fortunate enough to enjoy the best of an expensive industry.
I bet Young suffers hearing loss but what he’s talking about is training the brain and the problems with poor qulaity data. It’s about data loss with compression, in this case reducing a format designed to sample the entire audio spectrum 44,100 times per second into a paltry 3MB mp3.
btw I’m no audiophile and am someone who made a conscious decision years ago to risk decades of ear training for convenience in order to tolerate poor quality mp3 audio books at a shitty job!
2 Likes
Neil Young has old people Internet problems. Just like me, when I was at first elated to hear youngsters sing “Bella Ciao”, and then seriously butt hurt when I found out it had been turned into a depoliticized dance floor hit.
3 Likes
There are problems related to the transport of the final product and the production and mastering. Low bitrate DAB radio is hard to listen, a weak FM radio is more usable. Then there is the mastering process, that with digital techniques with CD and compressed format could be compressed in the audio meaning, reducing the dynamic range. This alone could make a recording dull and flat. With records to make them sound loud was possible spacing the grooves more.
But the elephant in the room is that using DAW and effect, especially the autotune and the time stretch or the click trank makes the master recording mechanical and not natural.
Older recording, made old style on a 24 track tape with all musicians playing in the same studio feel more natural, newer recording made with virtual instruments and editing software sounds synthetic and fake.
His ears and heart and memory are hearing what his music sounded like when he recorded it. He made this with his own hands, and his experience is part of his perception.
You and I will never hear what he has heard in his head and with his feelings. It doesn’t matter what medium any of us uses now (and I’m sure he has the best equipment out there), he will always hear things - imaginary or not - that you and I cannot.
The artist always has a different take on his own work than does his audience. If we were a race of empaths, perhaps we could experience what he does in his memory. Sadly, we are not.
5 Likes
Young appears not only to believe that the quality of the music is poor, which few would argue in comparison to vinyl
Um, excuse me? Vinyl is objectively worse than digital.
6 Likes
Yes, all of these points! You don’t have to be an audiophile or have magic ears to hear the difference between a CD and SiriusXM or free-tier streamed audio. If it’s just background music then it doesn’t really matter. But if I’m going to actually sit and listen to the music, I’ll go through the effort to pop in a CD, and play it through something significantly better than earbuds or PC speakers.
5 Likes
Sound goes in. Sound comes out. Never a miscommunication. You can’t explain that.
2 Likes
I think a problem I have with these discussions is that the association of “audiophile” with “person who buys a jar of magic stones to put on top of their speaker and shells out for activated oxygen-dosed cables” that was the tone of popular discussion of the movement a while back is still obnoxiously tenacious in my mind.
4 Likes
I was just using the term to mean roughly “someone that goes out of their way to achieve better quality sound reproduction”. How they go about accomplishing that, or to what extent, is at their own discretion. I do prefer to apply the term to those that can take physics into account and also understand the concept of diminishing returns.
3 Likes
Neil is an old crank, and his recent output is spotty at best, but he still makes some glorious noise. I’m glad he’s still managing to keep it weird.
4 Likes
I had a great post forming in my brain about the myth of the long lost Analog Augustan Age of vinyl, but since I got here a day late, others have beaten me to it and probably done it better.
I’ll just note that (1) nearly all of what’s bad about music today happens before the master is finalized (auto tune, robot drummers) and (2) “___________ hates what the internet has done to _________” is a great mad libs template.
5 Likes
Okay, no. Let’s unpack that a bit. “Live” music as in unprocessed music maybe just amplified from the original sound source and almost always far worse than ‘recorded’ music. It’s not the recording process (or any bit rate compression) that makes the difference, it’s the work that the mastering people do. They tweak volume levels, they adjust the tone, etc. Often they apply effects–think Phil Collins gated reverb. They do stuff that could never be done (by humans) to ‘live’ music. Artists can play multiple parallel tracks so that they can play/sing with themself. Think of songs like David Lee Roth’s “Skyscraper” where he had over 100 tracks overlaid. That’s not something you could do live.
So, people like to say ‘Live music sounds better than recorded’ when they mean that it sounds different and they life it sometimes. Because I can tell you, I’ve heard plenty of crap live music. Being live didn’t make it any better. It actually made it worse. If it had been recorded, I could have paused or to lowered the volume. “Live music is better” is just a form of nastalga. Yes, recording something and playing back doesn’t capture the original perfectly, but sometimes (and if it’s done right) it makes something much better. Think of special effects in movies. Being able to add to and manipulate a stored recording changed a whole industry. (several of them actually)
I think @philbin1 nailed it by saying that Neil compares recordings to what he heard in his brain when he made them and that they always fall short. Has everyone recorded themselves speaking or singing before? If you haven’t, I suggest you do it. You will be appaled by how bad you sound. Is that the recording’s fault? No, that’s what you sound like to other people. What you think you sound like is all in your head.
@Mind_is_Buddha I’m not sure what converstaion you’re having, but it’s clearly not the one the rest of use are.
@Elmer and @Mike.71 bring up a good point. Let’s be clear. There are two completely different uses of the term ‘compression’ in audio processing. One, that I think we can all agree is bad except in certain clever and limted (sorry, sorry) use cases is dynamic range compression. That is reducing the difference between the loudest part and quietest part of a passage. It’s that Phil Spector started to do in the 60’s and it’s haunted popular music since. I hope that’s not a topic for debate. The other meaning is ‘bit rate compression’ and that comes in two forms. Losless and lossy. In the former (FLAC, APE, something apple does) you get back exactly what you put in. (Sorry if I was a bit harsh on that @RandomDude) It’s completely unchanged. If you think otherwise, then there’s little I can do to help you. Lossy, on the other hand, is MP3 (Motion Pictures Expert Group One Layer Three Audio), Vorbis, AAC, AC3, Opus, etc. are all meant to reproduct the psycho-acoustical experience of a music recording. They take advantage of our understanding of how the whole sound to experience pathway in the human brain works. They remove things we can’t hear and/or don’t seem to care about. In doing so, they can decrease the amount of information that has to be stored to reproduce an audio signal that has a similar effect on the human auditory system as the original +/- some error. As you crank up the compression, that error increases. Generally speaking, the less error the better, but there are infrastructure limites to how little compression you can use. @GospelX touched on this. In general, there’s plenty of bandwidth on the internet to transmit perfect audio, but if you’re a low margin outfit like Spotify transmitting a million streams of audio, ever bit starts to cost real money, so they compress the heck out of it. That’s also why DAB and satellite sound so bad. There is only so much data they can transmit so they get what they get (and they often make poor decisions about what to transmit)
7 Likes
No worries, it’s a common misunderstanding. The kind you get when you take people who are used to using language exceedingly precisely (programmers) and mix that with actual humans. 
I’m not surprised that you found Vorbis to be (Vorbis is the audio codec, Ogg is a container format that lets you store multiple ‘streams’ of data like video+audio, multiple audio tracks, etc.) near CD quality even at reasonable bit rates (128Kbps). It’s very well designed and got to learn from a lot of mistakes made in MP3. Opus is even better, but some devices won’t support it because it’s ‘new’ (it’s almost 7 years old now).
If you want to do audio comparisons, I would suggest heading over to the Hydrogen Audio forums and reading a lot there. I do not hang out there, but I have often gotten search results that sent me there and I’ve been impressed with what I found. That may not be a fair sampling of all that’s there, but at least there is some signal in that noise. I believe they have a tool to help automate ABX blind testing which their members have used over the years to compare a lot of different audio processing things–wires, codecs, etc. They have had to suffer too much over the years with people who said “A is better than B” only to find out that it was the testing method that was wrong and that A was not better than B, so they’ve gotten very good with detecting (and preventing) that particular problem. And it’s a human thing to make that mistake–we all do it. But, fortunately, there are tools that can help us humans be better, so we should use them.
Sorry if I came off as abrasive, but someone was wrong on the internet.
And, to add to what @LDoBe said, lossless compression can be done by something like Zip, but please don’t use Zip for audio. FLAC (and others) work better because they know they’re dealing with audio and that audio has certain properties that random computer data doesn’t have, so they can do a better job.
@codinghorror Yep, as long as they’re using a reasonably good encoder, 128Kbps AAC should sound pertty much like the original source even on very good equipment. This is completely subjective, but Opus down to 80Kbps is fine for most audio I have on reasonable playback equipment (headphones and car), but I wouldn’t rely on it for a fixed audio system with good speakers. I’d want >100Kbps for that. Other codecs are slightly worse, so say 96Kpbs for AAC for mobile and 128 for fixed. Satellite/DAB at 32/48Kbps is just horrible.
5 Likes
It is a fucking abomination. I have no idea what they’re thinking, but I can’t listen to that, and no other human should be forced to, either.
4 Likes
